linux-next: build failure after merge of the vfs-brauner tree
Subsystems (2)
Mark Brown
maintainerHi all,
After merging the vfs-brauner tree, today's linux-next build (arm64 kselftest) failed like this:
CC statmount_test_ns
statmount_test_ns.c:37:19: error: static declaration of 'wait_for_pid' follows non-static declaration
37 | static inline int wait_for_pid(pid_t pid)
| ^~~~~~~~~~~~
In file included from statmount_test_ns.c:17:
../utils.h:47:12: note: previous declaration of 'wait_for_pid' with type 'int(pid_t)' {aka 'int(int)'}
47 | extern int wait_for_pid(pid_t pid);
| ^~~~~~~~~~~~make[1]: *** [../../lib.mk:225: /tmp/next/arm6
Caused by commit
32f54f2bbccfd (selftests/filesystems: add tests for empty mount namespaces)
I have used the version from next-20260312 instead.
Mark Brown
maintainerQuoted 9 lines · Mark Brown
Hi all,
After merging the vfs-brauner tree, today's linux-next build (arm64 kselftest) failed like this:
CC
statmount_test_ns
statmount_test_ns.c:37:19: error: static declaration of 'wait_for_pid' follows non-static declaration
37 | static inline intwait_for_pid(pid_t pid)| ^~~~~~~~~~~~
This issue is still present in today's -next, the version of the vfs-brauner tree being merged is currently that from next-20260312.
Quoted 6 lines · Mark Brown
Quoted 5 lines · Mark Brown
Hi all,
After merging the vfs-brauner tree, today's linux-next build (arm64 kselftest) failed like this:
Quoted 7 lines
Quoted 4 lines
CC
statmount_test_ns
statmount_test_ns.c:37:19: error: static declaration of 'wait_for_pid' follows non-static declaration
37 | static inline intwait_for_pid(pid_t pid)| ^~~~~~~~~~~~This issue is still present in today's -next, the version of the vfs-brauner tree being merged is currently that from next-20260312.
Sorry about this. Fixed now by appended patch.
From 418d1c0cdd872493731cf1f10f16a54b38a9ef57 Mon Sep 17 00:00:00 2001
From: Christian Brauner <[email protected]>
Date: Mon, 23 Mar 2026 16:31:30 +0100
Subject: [PATCH] selftests/statmount: remove duplicate wait_for_pid()
Remove the local static wait_for_pid() definition from statmount_test_ns.c as it conflicts with the extern declaration in utils.h. The identical function is already provided by utils.c.
--- .../filesystems/statmount/statmount_test_ns.c | 25 ------------------- 1 file changed, 25 deletions(-)
tools/testing/selftests/filesystems/statmount/statmount_test_ns.c +0 -25
@@ -34,31 +34,6 @@ static void handle_result(int ret, const char *testname)ksft_test_result_skip ("%s\n" , testname); } -static inline int wait_for_pid (pid_t pid) -{ -int status, ret; - -again: - ret =waitpid (pid, &status,0 ); -if (ret ==-1 ) { -if (errno ==EINTR ) - goto again; - -ksft_print_msg ("waitpid returned -1, errno=%d\n" , errno); -return -1 ; - } - -if (!WIFEXITED (status)) { -ksft_print_msg ( -"waitpid !WIFEXITED, WIFSIGNALED=%d, WTERMSIG=%d\n" , -WIFSIGNALED (status),WTERMSIG (status)); -return -1 ; - } - - ret =WEXITSTATUS (status); -return ret; -} -static int get_mnt_ns_id (const char *mnt_ns,uint64_t *mnt_ns_id) {int fd =open (mnt_ns,O_RDONLY );
Signature · 1 line
2.47.3
Mark Brown
maintainerQuoted 10 lines · Christian Brauner
Quoted 2 lines · Mark Brown
This issue is still present in today's -next, the version of the vfs-brauner tree being merged is currently that from next-20260312.
Sorry about this. Fixed now by appended patch.
Remove the local static
wait_for_pid()definition fromstatmount_test_ns.cas it conflicts with the extern declaration inutils.h. The identical function is already provided byutils.c.
Thanks, that looks like it should do it - I didn't see an update to the tree today though, did you forget to push or something?
Quoted 6 lines · Mark Brown
Quoted 4 lines · Christian Brauner
Quoted 2 lines · Mark Brown
This issue is still present in today's -next, the version of the vfs-brauner tree being merged is currently that from next-20260312.
Quoted 8 lines
Quoted 5 lines
Sorry about this. Fixed now by appended patch.
Remove the local static
wait_for_pid()definition fromstatmount_test_ns.cas it conflicts with the extern declaration inutils.h. The identical function is already provided byutils.c.Thanks, that looks like it should do it - I didn't see an update to the tree today though, did you forget to push or something?
The test-suite hadn't finished so I didn't want to risk it. Pushing in a bit.